{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:R5CRR3BTRYIGVXXDMHS6YKTZPZ","short_pith_number":"pith:R5CRR3BT","schema_version":"1.0","canonical_sha256":"8f4518ec338e106adee361e5ec2a797e51c1469e5ea71e1843dc17f797573039","source":{"kind":"arxiv","id":"2004.00584","version":3},"attestation_state":"computed","paper":{"title":"Deep Entity Matching with Pre-Trained Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.DB","authors_text":"AnHai Doan, Jinfeng Li, Wang-Chiew Tan, Yoshihiko Suhara, Yuliang Li","submitted_at":"2020-04-01T17:14:10Z","abstract_excerpt":"We present Ditto, a novel entity matching system based on pre-trained Transformer-based language models. We fine-tune and cast EM as a sequence-pair classification problem to leverage such models with a simple architecture. Our experiments show that a straightforward application of language models such as BERT, DistilBERT, or RoBERTa pre-trained on large text corpora already significantly improves the matching quality and outperforms previous state-of-the-art (SOTA), by up to 29% of F1 score on benchmark datasets. We also developed three optimization techniques to further improve Ditto's match"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2004.00584","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DB","submitted_at":"2020-04-01T17:14:10Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"782f62e9a67d5412778cc522a346957aa4721263d2dcaec3897db371817a1a76","abstract_canon_sha256":"a9c40329dcbcd15f91aa04dbb68946b4e35655adc2dbf6d8ea90420165f8c179"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:32:43.768819Z","signature_b64":"kGRHEJkWKK4fNF5dQq/b7da9aRMIrQ7wN+ugSD7JVyuc4vVFtPzRlRtuC5YVl7/AQrnrg2BSCvbb8Pe+ReQiAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8f4518ec338e106adee361e5ec2a797e51c1469e5ea71e1843dc17f797573039","last_reissued_at":"2026-07-05T01:32:43.768344Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:32:43.768344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Deep Entity Matching with Pre-Trained Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.DB","authors_text":"AnHai Doan, Jinfeng Li, Wang-Chiew Tan, Yoshihiko Suhara, Yuliang Li","submitted_at":"2020-04-01T17:14:10Z","abstract_excerpt":"We present Ditto, a novel entity matching system based on pre-trained Transformer-based language models. We fine-tune and cast EM as a sequence-pair classification problem to leverage such models with a simple architecture. Our experiments show that a straightforward application of language models such as BERT, DistilBERT, or RoBERTa pre-trained on large text corpora already significantly improves the matching quality and outperforms previous state-of-the-art (SOTA), by up to 29% of F1 score on benchmark datasets. We also developed three optimization techniques to further improve Ditto's match"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2004.00584","kind":"arxiv","version":3},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2004.00584/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2004.00584","created_at":"2026-07-05T01:32:43.768407+00:00"},{"alias_kind":"arxiv_version","alias_value":"2004.00584v3","created_at":"2026-07-05T01:32:43.768407+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2004.00584","created_at":"2026-07-05T01:32:43.768407+00:00"},{"alias_kind":"pith_short_12","alias_value":"R5CRR3BTRYIG","created_at":"2026-07-05T01:32:43.768407+00:00"},{"alias_kind":"pith_short_16","alias_value":"R5CRR3BTRYIGVXXD","created_at":"2026-07-05T01:32:43.768407+00:00"},{"alias_kind":"pith_short_8","alias_value":"R5CRR3BT","created_at":"2026-07-05T01:32:43.768407+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24407","citing_title":"Entity Resolution via Batched Oracle Queries","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25814","citing_title":"Adaptive Graph Refinement and Label Propagation with LLMs for Cost-Effective Entity Resolution","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02477","citing_title":"Guideline2Graph: Profile-Aware Multimodal Parsing for Executable Clinical Decision Graphs","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ","json":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ.json","graph_json":"https://pith.science/api/pith-number/R5CRR3BTRYIGVXXDMHS6YKTZPZ/graph.json","events_json":"https://pith.science/api/pith-number/R5CRR3BTRYIGVXXDMHS6YKTZPZ/events.json","paper":"https://pith.science/paper/R5CRR3BT"},"agent_actions":{"view_html":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ","download_json":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ.json","view_paper":"https://pith.science/paper/R5CRR3BT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2004.00584&json=true","fetch_graph":"https://pith.science/api/pith-number/R5CRR3BTRYIGVXXDMHS6YKTZPZ/graph.json","fetch_events":"https://pith.science/api/pith-number/R5CRR3BTRYIGVXXDMHS6YKTZPZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ/action/storage_attestation","attest_author":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ/action/author_attestation","sign_citation":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ/action/citation_signature","submit_replication":"https://pith.science/pith/R5CRR3BTRYIGVXXDMHS6YKTZPZ/action/replication_record"}},"created_at":"2026-07-05T01:32:43.768407+00:00","updated_at":"2026-07-05T01:32:43.768407+00:00"}